The efficacy of cisapride; time to review not regurgitate.

Abstract

In light of recent reports of adverse cardiovascular outcome in some patients taking cisapride, pediatric gastroenterology working groups from Europe and North America, have published recommendations on the use of this drug (J Pediatr Gastroenterol Nutr 1999;28:518–528; J Pediatr Gastroenterol Nutr 1999;28:529–533). The authors are to be congratulated on their in-depth reviews and their subsequent recommendations on the safety of this widely used prokinetic agent. However, we strongly believe that to document the implication of cisapride toxicity is meaningless in the absence of a systematic, critical overview of the efficacy of the drug in the first place. In both consensus statements, the authors indicate that they have reviewed the current literature regarding the efficacy of cisapride in gastroesophageal reflux in children. Both groups reference a number of studies and overviews of the use of this drug and state that their evaluation of the current literature suggests that this drug is effective in the setting of gastroesophageal reflux in children. However, we believe that it is inadequate simply to assert this opinion (re-stated by Nurko in a subsequent commentary J Pediatr Gastroenterol Nutr 1999;29:7) without careful substantiation through systematic evaluation of the available evidence. In contrast to the assertions contained in the ESPGHAN and NASPGN consensus statements, we feel it is at least as valid to interpret the current literature as not supporting the effectiveness of cisapride in gastroesophageal reflux. If we are to achieve a credible consensus in this controversial area, open and critical examination of the current literature will be essential. This is especially so given recent published evidence (J Pediatr Gastroenterol Nutr 1997;25:499–506; J Pediatr 1999;134:287–292) that casts significant doubts on the efficacy of cisapride for children with gastroesophageal reflux. In particular Scott et al. (J Pediatr Gastroenterol Nutr 1997;25:499–506), in the largest, randomized, double-blind trial of cleapride to date, could show no differences between cisapride and placebo groups for clinical parameters of reflux, manometric measurements or the majority of pH data. Moreover, in contrast to a previous study using a smaller number of children (Arch Dis Child 1987;62:454–457), there was no difference between treated and untreated groups with regard to improvement of esophagitis. In a more recent publication from Australia, Cohen et al. (J Pediatr 1999;134:287–292) demonstrated an improvement in some pH parameters among children on cisapride versus those on placebo. However, there was no difference in a number of pH scores, including the reflux score, and cisapride was no better than placebo for relief of vomiting. Indeed, it is instructive to scrutinize earlier studies that are often quoted in support of the utility of cisapride in children with gastroesophageal reflux. For instance, although the small, randomised, controlled trial of Cucchlara et al. (Arch Dis child 1987;62:454–457) showed both an improvement in histological score of oesophageal biopsies, and the overall clinical score over base-line in the cisapride group, there were no significant differences between the placebo and treatment groups with regard to any of these parameters. Similarly, Vandenplas (J Pediatr Gastroenterol Nutr 1991;12:44–47) could show significant differences only in the number of reflux episodes lasting longer than five minutes when comparing cisapride with placebo among twenty nine infants aged 2–4 months old with gastroesophageal reflux. There is little doubt that the aforementioned studies, and the plethora of other controlled and uncontrolled studies of cisapride treatment in childhood, are all open to criticism of one form or another in terms of the validity of the findings and the strength of the evidence either for or against the efficacy of cisapride in gastroesophageal reflux. In an enlightened editorial accompanying the most recent trial of cisapride in children with gastroesophageal reflux, Schulman (J Pediatr 1999;134:262–263) outlines many of the issues that hamper the interpretation of previous studies. These include the lack of distinction between those with gastroesophageal reflux (GER) and those with gastroesophageal reflux disease (GERD), the sample size and statistical analysis used, and the intrinsic biological variability of pH probes. We are in agreement with Schulman when he states “If, the drug is useful then we can take on the challenge of asking whether its benefits outweigh its potential cardiac risk” (our italics).